A CREATIVE COGNITION 211of the brain. And it is doubtful whether thinking really consists of se-
quences of simple if- then rules. Th at assumes a repetitiveness of experi-
ence that is rare in the real world. So the rules of real cognition must be
of a diff er ent form— ones that regularly need to be updated.
Th is may explain why the implementations of computational models
in artifi cial intelligence have produced machines that do simple, routine
things (like select and count items according to fi xed criteria) very well.
But they cannot deal with fuzzy, dynamic inputs like moving images or
human speech, except by breaking down such inputs into simpler steps.
Discrete characters and steps suit a computer program, but this is not
what real experience is like. So those programs are unlikely to yield truly
realistic models of thinking.
Similarly, for the purposes of cognitive research, very simple, well-
defi ned, stationary (nondynamic) tasks devoid of context have been re-
sorted to. But the strategy has prob ably been self- defeating, because most
everyday prob lems are dynamical, ill defi ned, and intimately related to
context. Consequently, we seemed to have learned a lot about the diffi -
culties humans have with unreal cognitive tasks in an unreal world, but
not so much about real thinking in the real world.
Th e strategy has also encouraged a remarkable underestimation of
human thinking/reasoning powers in the real world. Th e pessimism
about the cognitive ability of both adults and children in cognitive test-
ing (as in IQ testing) was mentioned earlier. In contrast, studies of prob-
lem solv ing in rea l- life contexts revea l complex ities of t hink ing t hat g reat ly
exceed those manifested in such tests (see chapter 3).
We have prob ably also underestimated the cognitive abilities of non-
human animals.^10 Research over the past few years has revealed previ-
ously unsuspected complexities of cognitive pro cessing in invertebrates
such as bees and fl ies, as well as vertebrates, such as fi sh and jackdaws.^11
It seems clear that current models of thinking are failing to capture the
true qualities of cognitive operations in complex, dynamic situations,
from the foraging of the honeybee to humans carry ing out cooperative,
constructive acts over long periods.
In sum, a vast variety of work is going on in a wide variety of special
cognitive topics. But the attainment of an agreed-on model of what
This content downloaded from 139.184.14.159 on Tue, 17 Oct 2017 13:54:24 UTC